Technical Field
The present invention relates to cloud service computing environments and more particularly to systems and methods with reduced latency time in using cloud network services including continuity management, testing and other services.
Description of the Related Art
Business Continuity Management (BCM) aims at sustaining an organization's business operations without disruption. Disaster Recovery Planning (DRP) is a subset of BCM that focuses on preparing for and recovering from disasters that threaten to seriously affect information technology (IT) service availability and thereby to disrupt business continuity. IT Service Continuity Management (ITSCM) formalizes DRP from an IT Service Management (ITSM) perspective. The IT Infrastructure Library (ITIL), a widely recognized framework of ITSM best practices, describes the ITSCM lifecycle stages: Initiation, Requirements and Strategy, Implementation, On-Going Operation and Invocation.
ITSCM, particularly in distributed systems, is widely perceived as an expensive challenge for enterprise-class IT operations. The daunting complexity of resource management makes labor cost a significant factor in pursuing continuity and resiliency of a business. Predominantly small and medium businesses (SMB) are known to opt out from ITSCM due to high cost to the business in labor, hardware resources and the high price of services of external service providers. The medium and large enterprises who subscribe to ITSCM services, on the other side, fail to maintain currency of continuity plans and neglect frequent rehearsals due to high costs involved and potentially negative influences of rehearsals on production services. This failure leaves enterprises poorly prepared and jeopardizes the chances of rapid recovery in the event of a real disaster.
Bare-Metal Recovery allows backing up a computer system from and restoring it to a physical infrastructure. Flash Archive™ is a utility to clone and secure a computer system to disk or tape volumes. Hewlett-Packard® UX provides Ignite-UX™, a utility to clone and secure a computer system to network, disk or tape volumes. IBM® AIX™ offers mksysb, savevg and other utilities to clone and secure a computer system to network, disk or tape volumes. Linux based Intel® systems can use live backup mechanisms on the running system, such as file system snapshots based on logical representations of the data using the Logical Volume Manager (LVM). As an alternative to live backups, the backups can also be performed with support of the “alternate boot” method, i.e., a backup of the whole image requires rebooting the system. Windows® based Intel® systems support similar “alternate boot” backup and restore mechanisms.
All Bare-Metal Recovery techniques have a significant disadvantage: they require a similar physical infrastructure to recover on. This requirement can be fulfilled by setting up stand-by redundant physical infrastructure on a recovery site. Such a redundant site is usually provided as a service by an ITSCM service provider, and comes with a high price tag. One can further differentiate between cold, warm, hot and shared redundancy. Cold redundancy is based on an active/passive scenario with replicated hardware that must be set up and configured before it is usable. Warm redundancy is based on an active-passive scenario with pre-configured replicated hardware that must be manually brought into the system, whereas hot redundancy uses automated failover mechanisms in an active-active scenario. Shared redundancy relies on the simultaneous use of the production hardware and the replicated hardware.
Virtualization technology can be used to combine the recovery of computer systems with server consolidation, i.e., recovering physical servers on Virtual Machines (VM) that operate on a smaller pool of physical infrastructure resources. Furthermore, VM live migration and replication offers approaches to automate availability planning and ITSCM by moving workloads without service disruption. Mirage™ proposes an approach to improve inventory control of VM disk-images and fight image sprawl. One important idea of Mirage™ is to enrich VM images with metadata in a manifest that allows the separation of the storage of a VM and operations on it, such as search and update without having to start the VM.
However, these solutions introduce hypervisor dependencies and thereby might lead the customer into a technology lock-in situation. Although efforts are made to enhance hypervisor interoperability, such as the Open Virtualization Format, it is unclear how these standards can be effectively incorporated into higher-level business processes.
Recovery to virtual infrastructure can reduce the up-front cost of redundant physical hardware. However, the cost savings are limited and stand-by redundant hardware must still be provided by the customer or by an ITSCM service provider.
Cloud Computing provisions Internet-scale services which are mainly characterized by three properties: they provide means for on-demand mobilization of infrastructure resources, they exhibit the illusion of infinite amounts of resources being available, and they implement the idea of utility computing. The interest in cloud computing has been fueled by innovative Web service offerings. Customers face a number of challenges when outsourcing systems “into the cloud”. For example, building secure virtualized networking solutions to complement simple cloud services to design more complex enterprise services. Other challenges come from the before-mentioned hypervisor interoperability problems. For example, VMs created with VMware cannot be brought into a system which uses VMs based on the Xen hypervisor.